Complexity of Mycobacterium avium complex immune reconstitution inflammatory syndrome (MAC-IRIS) in a patient with HIV

Abstract

We report a case of Mycobacterium avium complex immune reconstitution inflammatory syndrome (MAC-IRIS) in a patient with HIV positive. Initial presentation was that of a purpuric purple macular rash in-keeping with Kaposi sarcoma as an AIDS defining illness. Three weeks following the initiation of antiretroviral treatment (ART) she developed chest pain, dry cough and fever. A diagnosis of MAC was made through imaging and sputum cultures and appropriate treatment was initiated. Despite adequate management with evidence of good immunological and virological response, the patient represented with persistent symptoms. Repeat CT of the chest confirmed worsening lymphadenopathy with necrosis. Given these findings, a diagnosis of MAC-IRIS was made with resolution of fever after corticosteroids were initiated. This case highlights the importance of considering MAC as a cause of IRIS in severely immunosuppressed patients with HIV.

Background

Immune reconstitution inflammatory (IRIS) in HIV-infected patients initiating antiretroviral treatment (ART) results from restored immunity to a specific infectious or non-infectious agent.¹ It is characterised by a paradoxical clinical worsening of a known condition or the development of a new condition due to dysregulation of the immune system to both intact subclinical pathogens and residual antigens.² Clinical deterioration is thus seen in these patients with worsening or appearance of new symptoms or radiographic manifestations of opportunistic infections after initiation of ART. This is not the direct result of treatment failure or another disease process.

Although IRIS is an extensively researched condition with tuberculosis-IRIS (TB-IRIS) being the most common opportunistic infection mentioned in the literature, Mycobacterium avium complex (MAC)-IRIS is not frequently reported and to date, no trials concerning the management of MAC-IRIS have been conducted. Evidence regarding treatment is limited to small case reports or case series which report their management practice and thus do not provide reliable evidence regarding the safety or efficacy of their approach. MAC-IRIS may be self-limiting but may require adjunctive therapy in severe cases akin to our patient. We report a case of MAC-IRIS in an HIV-infected patient as a reminder of an important clinical lesson, highlighting the complexity of diagnosis based on clinical presentation and exclusion of alternative causes of deterioration and management of such a condition.

Case presentation

Our case focuses on a 28-year-old Colombian woman who had lived most of her life in the UK and presented to our clinic in Malta with a 1-year history of a pruritic violaceous macular rash over trunk and limbs, not responding to topical corticosteroids. She also reported a 6-month history of weight loss, anorexia and rigours but no documented fever.

She was otherwise well and was on no regular medication. Her medical and family history were unremarkable. She was in a stable relationship with her male partner for the last 2 years and denied any alcohol intake, smoking or illicit drug use. She had been residing in Malta for the last 9 months and was employed as a waitress. There was no significant travel history of note except for a family visit to Colombia a few years ago.

On clinical assessment, the patient was found to be afebrile but tachypnoeic at 24 breaths/min and tachycardic at 110 beats/min. Examination revealed oral candidiasis and a rash consistent with Kaposi sarcoma (figure 1). Full physical examination was otherwise normal.

Figure 1.

Violaceous macular rash consistent with Kaposi sarcoma.

In view of the presenting features, the patient was consented for an HIV test which was confirmed to be positive. She was therefore started on ART and discharged a few days later.

Three weeks into treatment, the patient presented again with persistent low-grade fever, worsening fatigue associated with a dry cough, intermittent chest discomfort and palpitations. Chest X-ray (CXR) showed new perihilar shadows. Hospitalisation for further investigations revealed positive sputum cultures for non-tuberculous Mycobacteria (NTM) MAC by PCR and the appropriate treatment was initiated.

Subsequently, the patient presented again with symptoms of persistent fever and worsening symptoms not responding to antipyretics despite full adherence to her optimised ART and MAC treatment. In addition, she also reported a feeling of ‘fullness in the chest’ which was exacerbated on lying down.

Investigations

Initial blood investigations revealed a lymphopenia (lymphocytes 0.46×10⁹/L) and mildly raised inflammatory markers (C reactive protein 39 mg/L). As part of the workup for anaemia, leishmania antibodies and parvovirus PCR were taken but not detected. HIV test was positive—the initial CD4 count was 10 cells/mm³ with an HIV viral load of >10 000 copies/mL.

A baseline CXR showed perihilar shadowing (figure 2) and a follow-up CT scan of the thorax showed increase in size of the bilateral hilar and superior mediastinal lymphadenopathy with necrosis and excluded thromboembolic disease (figure 3). The imaging was reviewed by a respiratory radiologist and a CT guided lung biopsy was suggested.

Figure 2.

(A) Baseline chest X-ray (CXR) showing bilateral perihilar shadowing and widening of the right paratracheal stripe. (B) Repeat CXR on second presentation, 3 weeks after showing interim increase in the bilateral hilar enlargement with right apical consolidation.

Figure 3.

(A) Initial CT thorax showing bilateral hilar and superior mediastinal lymphadenopathy, the majority of which demonstrate features of necrosis. (B) Repeat CT thorax 3 weeks after presentation showing an increase in size of the previously described bilateral hilar and mediastinal lymphadenopathy.

Investigations for other potential opportunistic infections such as serology for Epstein-Barr virus, toxoplasma, cytomegalovirus and cryptococcal antigen testing resulted as negative. A blood born viral screen (including Hep B and C) was also negative. Fungal markers including beta-D-glucan and galactomannan were also negative. Quantiferon test was not reactive.

Two out of three sputum samples smears tested positive for Ziehl-Neelsen staining with grade 2+ acid-fast bacilli (AFB), and NTM MAC was subsequently cultivated by PCR. Given that the diagnosis of MAC infection was now confirmed through sputum culture, a lymph node biopsy was not pursued.

Differential diagnosis

This case highlights the complexity in investigating and managing patients with HIV. The patient was diagnosed with Kaposi sarcoma as an AIDS-defining illness. A panel of blood and sputum investigations including serology and PCR testing were taken to exclude the most common opportunistic infections. Pneumocystitis jirovecii pneumonia (PCP) was initially suspected but eventually ruled out when a contrast CT thorax confirmed that the changes were not intraparenchymal. PCR on nasopharyngeal swab was also negative for PCP.

The differential diagnosis of necrotic lymphadenopathy included tuberculous mycobacterial infection, non-tuberculous mycobacterial infection, fungal infections and lymphoproliferative disorders such as lymphoma. Histological sampling through biopsy would have been ideal for immunohistochemical staining to exclude lymphoma or malignancy, but this was deemed unnecessary once a diagnosis of NTM-MAC by PCR from liquid culture was confirmed.

The working diagnosis of a mycobacterial infection was confirmed with two positive sputum samples for AFBs. MAC, sensitive to macrolide antibiotic treatment, was identified on whole genome sequencing and treatment was tailored accordingly.

Despite targeted antimicrobials, the patient still reported worsening of symptoms and remained febrile. At the first instance, a working diagnosis of possible treatment failure or drug hypersensitivity was made.

After 6 weeks of ART, a significant decrease in HIV viral load of 109 copies/mL and a rise in CD4 count of 106 cells/mm³ were noted. A diagnosis of MAC-IRIS was made, based on the typical presentation of opportunistic infection in a patient responding to ART with a precipitous drop in HIV viral load in a short period of time.

Treatment

The patient was started on tenofovir disoproxil fumarate 245 mg/day, lamivudine 150 mg 12 hourly and raltegravir 400 mg 12 hourly during her first presentation. Trimethoprim/sulfamethoxazole (Septrin) 960 mg/day was also started for PCP prophylaxis in view of a CD4 count of less than 200 cells/mm³. A 10-day course of fluconazole 400 mg/day was also prescribed for oral candidiasis.

Initial anti-TB treatment consisting of rifinah 150 three tablets a day (rifampicin and isoniazid), ethambutol 15 mg/kg and pyrazinamide 1.5 g/day was switched to rifabutin 300 mg/day, ethambutol 15 mg/kg and clarithromycin 500 mg 12 hourly once NTM was confirmed, based on sensitivities.

The persistence of symptoms despite adequate MAC treatment and adherence to ART was attributed to a diagnosis MAC-IRIS and she was started on oral prednisolone 40 mg/day, which was tailed down by 5 mg every 5 days.

Outcome and follow-up

The patient improved significantly with resolution of fever and chest symptoms over the first week of steroid therapy and was discharged after a week with close community and outpatient follow-up at the infectious disease clinic. Adherence to ART and MAC treatment including Septrin prophylaxis was emphasised, and both the patient and partner were counselled about HIV transmission prior to discharge. The partner tested negative.

HIV viral load was undetectable after 3 months of ART and the CD4 count increased to 100 cells/mm³. A repeat chest radiograph 2 months after initiating MAC therapy showed partially resolving lymphadenopathy with a negative sputum smear. The patient is being reviewed regularly, with the aim to treat the pulmonary infection for a minimum of 1 year, with repeat imaging to assess resolution of lymphadenopathy and decide further course of treatment.

Discussion

In our case, the primary presenting features were a longstanding purple, purpuric macular rash, associated with a history of weight loss, anorexia and rigours. This was followed by cough, fever and chest pain. Despite a good virological response and gradual restoration of immune function on commencing ART and antimycobacterial regimen, later tailored to anti-MAC treatment, the patient still experienced persistent fever and worsening CT thorax findings. The persistence of symptoms, the precipitous drop in HIV viral load coupled with a relatively rapid elevation in CD4 count and exclusion of other potential causes, lead us to the diagnosis of MAC-IRIS.

According to the British HIV Association (BHIVA) a low baseline CD4 cell count and a rapid recovery in CD4 numbers appear to be relevant in developing IRIS and does not appear to be associated with any particular ART regime or class.³ Possible risk factors for the development of IRIS as reported by a study by Shelburne et al⁴ include—short interval between initiating treatment for opportunistic infections and starting ART and rapid fall in HIV-1 RNA after ART. Other risk factors not relevant in our case include male sex and being ART-naïve at the time of opportunistic infection (OI) diagnosis. Further significant predictors mentioned in the literature include younger age, a lower baseline CD4 cell percentage, a lower CD4 cell count at ART initiation and a lower CD4 to CD8 ratio at baseline as in our case.⁵ Such conclusions regarding risk factors and incidence/prevalence of IRIS are difficult to ascertain as only small data sets are found in the literature and cohorts studied differ substantially on the type of IRIS examined (with most studies based on TB-IRIS only for example, which is the most common presentation).

Many atypical mycobacteria have been reported in HIV-infected patients with advanced immunosuppression, with most having disseminated focal disease. The most common of these infections reported are MAC and Mycobacterium kansasii.^{6 7}

The onset of fever in our case report was reported on day 21. Several reports have studied the time of onset of Mycobacterium-associated IRIS—in one study fever was the first sign of IRIS and occurred between 6 and 20 days after initiation of ART.² Other literature reports the median time interval from the start of ART to the development of mycobacterial lymphadenitis to be 17 days.⁸

To date, no randomised control trials concerning the management of IRIS have been conducted and thus there is a paucity of data to guide management in this condition. Continuation of ART remains of utmost importance, when possible. In severe cases where discontinuation of ART is considered, the potential disadvantages of therapy cessation, such as HIV progression or development of viral resistance, should be considered.¹ Corticosteroid treatment in severe IRIS seems to feature in most of the available literature on this uncommon condition. Starting such therapy in patients with severe immunosuppression can lead to a medical dilemma but this seems to be the cornerstone of treatment. However, there seems to be no consensus on the most appropriate corticosteroid regimen.

Antimicrobial treatment of disseminated MAC requires combination therapy with a macrolide and ethambutol, with or without rifabutin as quoted by the BHIVA guidelines. Rifabutin is considered in cases with advanced immunosuppression such as in our case. A retrospective study by Riddell et al⁹ based on 20 cases of MAC-IRIS over 8 years in four different HIV clinics, reported that 80% of the study cohort had a complete response to continuation of ART and antimycobacterial agents over a 36-month follow-up period, with a tendency of responders to maintain a more substantial CD4 cell count increase than non-responders.

Mild cases of IRIS may respond to symptomatic therapy alone, with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) as an acceptable treatment option in the setting of opportunistic infections.¹⁰ More severe cases of IRIS associated with mycobacterial and fungal infections require corticosteroid therapy. As demonstrated in our case, corticosteroid therapy with 20–40 mg of oral prednisolone a day for 4–8 weeks has been shown to have anecdotal benefit in IRIS according to the literature.¹¹ Other management with anecdotal benefit particularly in MAC-IRIS include the use of interleukin 2 and granulocyte-macrophage colony-stimulating factor which were found to be successful in a small cohort¹² and leukotriene inhibitors which were used in TB-associated IRIS in cases refractory or intolerant to steroids.¹³ Occasionally, surgical excision or repeated needle aspiration of profoundly enlarged lymph nodes or debridement of necrotic areas is reported, as used with Mycobacterial TB lymphadenitis, however, healing is often poor.¹⁴

In conclusion, MAC-IRIS should be a plausible differential whenever new lymphadenopathy coupled with fever arises in a recently treated patient with HIV with ART. Distinguishing between TB or MAC-IRIS clinically and radiologically can be very challenging. Identifying between the two is crucial as management differs. Nonetheless, new modalities in diagnostics (eg, PCR and Whole Genome Sequencing) can expedite the microbiological diagnosis and thereby offering appropriate treatment.

Learning points.

• Immune reconstitution inflammatory syndrome (IRIS) is a diagnosis of exclusion and is commonly misdiagnosed as HIV progression or failure of treatment.

• A diagnosis of IRIS should be considered in an HIV-infected patient in whom a good viral and immunological response to antiretroviral treatment (ART) is evident, however, subsequent deterioration in symptoms is present—with fever usually being the first sign.

• Although most of the HIV-IRIS reported in the literature is tuberculosis-related, this should also be considered in non-tuberculous Mycobacteria.

• Management of Mycobacterium avium complex-IRIS is still debatable given the lack of data and clinical trials available to date. Recommended treatment includes continuation of ART, in addition to a macrolide and ethambutol, with or without rifabutin.

• Corticosteroids remain the cornerstone of treatment for IRIS. However, more research is needed to confirm best steroid regimes in similar scenarios.

Ethics statements

Patient consent for publication

Obtained.